Presenter : Ricky Kerndt
Major : Computer Science
Poster 35
“Intra/Inter-network traffic has become an important part of our daily lives. Its become a primary means of communication through email, messaging, and social networks (e.g. Facebook, Twitter). The University of Oregon represents a small community environment encompassing student housing and the daily activity of classes, staff and professors carrying out research, administration and services to keep the campus functioning. The University’s network backbone thus provides a potential data source for studying how society uses internet applications in their daily activities. This project evaluates the potential of using anonymized netflows obtained from the UO Network and Telecom Services (NTS) to characterize network activity. Netflow records are provided with local addresses anonymized from UO border routers to preserve confidentiality. The records are aggreagated and stored in a database for later characterization of network activity. The characterization includes packet rate!
s, bandwith utilization, applications (unique ports), and distribution of destination IP address. We can then look at how this charac- terization differs with different areas of campus (dorms, offices, wifi) and temporal patterns. The results show that netflow records will provide a valuable data source for studying how a community setting uses internet applications in carrying out their daily activities.